During its Ignite 2020 conference, which kicked off virtually this morning, Microsoft announced updates to Azure Cognitive Services and Azure Machine Learning aimed at streamlining business processes during the coronavirus pandemic. The company also launched two features in Azure Cognitive Search — Private Endpoints and Managed Identities — plus enhancements to Bot Framework Composer and the broader Azure Bot Service.

“We’re seeing AI touching every business across the planet, and so one of the key focuses we have with Azure Machine Learning is to provide our customers with the tools to really simplify the ability to create new models because we know they’re going to need them in every area of their business,” Microsoft corporate vice president Eric Boyd told VentureBeat in a phone interview. “This continues to be a key theme for us — how we will really help our customers, enable more of their developers, and even more of their data analysts to build machine learn models and apply them in all aspects of their business.”

Cognitive Search

Private Endpoints in Cognitive Search, which is generally available as of today, allow a client on a virtual network to access data in an index over a private link. (Private links are provided via Azure Private Link, a paid service.) The private endpoint uses an IP address from the network for the search service. Network traffic between the client and search service traverses the network and private link, eliminating exposure from the public internet.

Also now available is Managed Identities, a feature that can create a data source object with a connection string that doesn’t include any credentials. The search service is granted access to the data source through role-based access control. When setting up a data source using a managed identity, credentials can be changed without impacting indexers’ ability to connect, and new objects can be generated without having to include or retrieve an account key.

Cognitive Services

New to Azure Cognitive Services in preview is Metrics Advisor, which helps developers embed data monitoring into apps and ostensibly makes it easier to monitor the performance of an organization’s growth engines while diagnosing potential issues. Microsoft says it is built on Anomaly Detector, its cloud-based service that looks at time-series data and automatically selects the right algorithm to maximize accuracy for scenarios like business incidents, monitoring IoT device traffic, managing fraud, and responding to changing markets.

Anomaly Detector is now generally available. Through an API, developers can use it to detect any level of anomaly — in the cloud or on edge devices with containers. Microsoft says Anomaly Detector, which the company uses internally across over 200 product teams, eliminates the need for labeled training data and requires no more than three lines of code.

In tow with Metrics Advisor is Spatial Analysis (also in preview), an Azure Cognitive Services feature that helps businesses create in-room layouts that support social distancing and other health compliance measures. It complements Store Traffic, a Dynamics 365 solution launched in July that gives store owners insights to help them remain below capacity limits and keep customers at safe distances.

Microsoft says Spatial Analysis was developed with “strict ethical standards and guidance” on how to implement it responsibly. “Applying responsible AI recommendations grounded in user and societal research, Spatial Analysis helps organizations optimize their space while maximizing personal privacy, providing transparency and promoting trust amongst their customers,” a spokesperson told VentureBeat via email.

As of June, Cognitive Services had tens of thousands of paying customers — 85% of the Fortune 100 — making 13.5 billion transactions, according to Microsoft.

Machine Learning

On the Azure Machine Learning side, Designer lets users visually connect datasets and modules on an interactive canvas to build, test, and deploy machine learning models. Designer supports connecting modules to create pipeline drafts and submitting pipeline runs using Azure Machine Learning workspace resources. (Pipelines, which consists of connected datasets and analytical modules, can be used to train a single model or multiple models, make predictions in real time or in batch, and clean data.) Training pipelines can be converted to inference pipelines and published to a pipeline endpoint to run with different parameters and datasets, and inference pipelines can be deployed to endpoints to make predictions on data in real time.

Automated ML UI is also now generally available in Azure Machine Learning. Based on innovations from Microsoft Research, Automated ML UI allows data scientists to build AI models more easily while sustaining high model quality. Developers can build and deploy predictive models for common use cases — such as classification, regression, and forecasting — without having to write code.

Dovetailing with Designer and Automated ML UI, ML Assisted Labeling triggers automatic machine learning models to accelerate data labeling, as well as a new set of security and enterprise readiness features for Azure Machine Learning. Operation-level role-based access control security support, currently in preview, offers granular control over AI projects to set custom roles or reuse prebuilt roles in order to control specific operations for individual users in a workspace. And Azure ML Integration with mlflow, an open source platform for AI lifecycle management, now includes support for submitting jobs to the cloud, model registry and deployment support, and expanded UI experimentation features.

Bot Service

Microsoft says Bot Framework Composer, an open source visual bot design and authoring tool, is now available through Virtual Assistant Solution Accelerator. Introduced in May, it provides a template for developing and embedding bots on channels like web chats, Facebook Messenger, Microsoft Teams, and Power Virtual Agent.

Lastly, the Alexa channel is now generally available within Azure Bot Service. It lets developers build a bot and deploy it across Alexa-enabled devices like smart speakers, smart displays, and apps. According to Microsoft, any Alexa device that supports custom skills is compatible.

Microsoft says over 2.5 billion messages have been exchanged using Bot Service.